WO2017202132A1 - 一种端口优化方法、装置及系统 - Google Patents

一种端口优化方法、装置及系统 Download PDF

Info

Publication number
WO2017202132A1
WO2017202132A1 PCT/CN2017/078287 CN2017078287W WO2017202132A1 WO 2017202132 A1 WO2017202132 A1 WO 2017202132A1 CN 2017078287 W CN2017078287 W CN 2017078287W WO 2017202132 A1 WO2017202132 A1 WO 2017202132A1
Authority
WO
WIPO (PCT)
Prior art keywords
port
adjustment value
bandwidth adjustment
bottleneck
convergence
Prior art date
Application number
PCT/CN2017/078287
Other languages
English (en)
French (fr)
Inventor
陆钱春
李锋
张祖红
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2017202132A1 publication Critical patent/WO2017202132A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability

Definitions

  • This document refers to, but is not limited to, the field of telecommunications and enterprise network management, especially a port optimization method, device and system.
  • the port bandwidth utilization gradually becomes reasonable.
  • the actual flow rate of all services in the network port does not reach the guaranteed bandwidth of the service at the same time.
  • the bandwidth of the network port is optimized to realize the rational use of network resources.
  • the traditional bandwidth optimization method can only roughly optimize the bandwidth of most ports, and cannot make finer optimization of the port bandwidth, and can not fully utilize network resources, thereby failing to ensure more effective multiplexing of local bandwidth of the network, thereby affecting users. The experience when using the web.
  • Embodiments of the present invention provide a port optimization method, apparatus, and system to fully utilize network resources.
  • the embodiment of the invention provides a port optimization method, including:
  • the convergence ratio of the bottleneck port is optimized according to the actual bandwidth adjustment value.
  • the determining a bottleneck port in the convergence port includes:
  • the bottleneck port in the convergence port is determined according to the bandwidth adjustment value and the bandwidth adjustment interval.
  • calculating an actual bandwidth adjustment value of the bottleneck port includes:
  • the convergence ratio optimization of the bottleneck port according to the actual bandwidth adjustment value includes:
  • the convergence ratio of the bottleneck port is optimized according to the actual bandwidth adjustment value
  • At least one service tunnel group is added, and the actual bandwidth adjustment value of the combined combination of all service tunnel groups is calculated until the actual bandwidth adjustment value satisfies the error analysis condition.
  • the convergence ratio optimization of the bottleneck port according to the actual bandwidth adjustment value includes:
  • calculating an actual bandwidth adjustment value of the bottleneck port includes:
  • the embodiment of the invention further provides a port optimization device, comprising:
  • a bottleneck port determining module configured to determine a bottleneck port in the convergence port, where the bottleneck port is a convergence port whose bandwidth adjustment value is outside the bandwidth adjustment interval;
  • a bandwidth calculation module configured to calculate an actual bandwidth adjustment value of the bottleneck port
  • the convergence ratio optimization module is configured to perform convergence ratio optimization on the bottleneck port according to the actual bandwidth adjustment value.
  • the bottleneck port determining module includes:
  • a bandwidth optimization sub-module configured to optimize the bandwidth of each port
  • a bandwidth selection submodule configured to select a bandwidth adjustment value of the convergence port from the optimization result
  • the bottleneck port determining sub-module is configured to determine a bottleneck port in the convergence port according to the bandwidth adjustment value and the bandwidth adjustment interval.
  • the bandwidth calculation module includes:
  • Finding a submodule configured to find a combination of service tunnel groups under the bottleneck port
  • the bandwidth calculation sub-module is configured to calculate an actual bandwidth adjustment value of the combination of the service tunnel group.
  • the convergence ratio optimization module includes:
  • a determining sub-module configured to determine whether the actual bandwidth adjustment value satisfies an error analysis condition, where the actual bandwidth adjustment value of the convergence port is within a convergence interval
  • a first convergence ratio optimization submodule configured to optimize a convergence ratio of the bottleneck port according to the actual bandwidth adjustment value when the actual bandwidth adjustment value satisfies an error analysis condition
  • the second convergence ratio optimization submodule is configured to add at least one service tunnel group when the actual bandwidth adjustment value does not satisfy the error analysis condition, and calculate an actual bandwidth adjustment value of the combination of all the added service tunnel groups until the The actual bandwidth adjustment value satisfies the error analysis condition.
  • the embodiment of the invention further provides a port optimization system, comprising: a port and a port optimization device as described above.
  • the embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the port optimization method of any of the foregoing.
  • the embodiment of the present invention provides a port optimization method, device, system, and computer storage medium.
  • the bottleneck port is a convergence port whose bandwidth adjustment value is outside the bandwidth adjustment interval, and the actual bottleneck port is calculated.
  • the bandwidth adjustment value is used to optimize the convergence ratio of the bottleneck port according to the actual bandwidth adjustment value.
  • the bandwidth of the bottleneck port is adjusted by the joint optimization of the port bandwidth and the convergence ratio, so as to ensure that the port convergence ratio, the guaranteed bandwidth, and the port utilization are dynamically balanced under the condition of sufficient resources, so that The port bandwidth is finerly optimized.
  • the joint optimization method further optimizes the convergence ratio based on the bandwidth optimization of all ports in the optimized range to more effectively utilize network resources.
  • FIG. 1 is a flowchart of a port optimization method according to Embodiment 1 of the present invention.
  • FIG. 2 is a schematic diagram of a port optimization apparatus according to Embodiment 3 of the present invention.
  • This embodiment provides a port optimization method. Referring to FIG. 1, the steps are as follows:
  • the system collects parameter information such as bandwidth, traffic, convergence ratio assets and performance data of all ports (including convergence ports and non-convergence ports) of a certain subnet resource in a certain period, and then according to selected parameter information and bandwidth optimization strategy.
  • the bandwidth of each port during this period is optimized.
  • the bandwidth optimization strategy is to provide an initial bandwidth adjustment value of the service tunnel under the port and an initial bandwidth adjustment value of the pseudowire carried by the service tunnel according to the collected parameter information.
  • the bandwidth adjustment value of the convergence port is selected from the optimization result, and the bottleneck port in the convergence port is determined according to the bandwidth adjustment value and the bandwidth adjustment interval.
  • the bandwidth adjustment interval is an interval determined by the bandwidth utilization and the port convergence ratio after the bandwidth optimization.
  • the bandwidth utilization can be manually configured, and the convergence ratio can also be manually configured through policies.
  • the bandwidth adjustment interval is an interval set according to the collected parameter information and the convergence ratio optimization strategy.
  • the convergence ratio optimization strategy is to seek a reasonable port convergence ratio and bandwidth utilization according to the acquired parameter information.
  • the bandwidth adjustment interval may also be set by the user according to the collected parameter information.
  • the foregoing bottleneck port refers to a port network resource shortage.
  • the port convergence ratio cannot be adjusted to provide sufficient bandwidth to configure corresponding service requirements.
  • the convergence port includes at least one bottleneck port. If the bottleneck port is not included in the convergence port, it is considered that each port with bandwidth optimization is sufficient, and the convergence ratio can be directly optimized without adjusting the bandwidth adjustment values.
  • the premise of the convergence ratio optimization of the bottleneck port port is that the port guarantees the bandwidth adjustment value, the port convergence ratio, and the bandwidth utilization balance.
  • the premise of the convergence ratio optimization of the bottleneck port port is that the port guarantees the bandwidth adjustment value, the port convergence ratio, and the bandwidth utilization balance.
  • the optimization of the convergence ratio is achieved by calculating at least one of the convergence ratio adjustment value and the bandwidth utilization adjustment value.
  • the optimization method for each port bandwidth may adopt an existing optimization manner, such as dynamically adjusting the bandwidth currently configured by the port according to the bandwidth required by the user to meet the user requirement, and solving the problem of insufficient port bandwidth remaining.
  • the bandwidth adjustment value of each port and the service tunnel bandwidth adjustment value in each port are obtained.
  • the port includes a convergence port and a non-convergence port, and then selects a bandwidth adjustment value of at least one convergence port from the port. Select the bandwidth adjustment value of all converged ports from the port.
  • the bandwidth adjustment interval of the convergence port is determined, and then the bandwidth adjustment value of the convergence port is determined to be outside the bandwidth adjustment interval. If the bandwidth adjustment value of the convergence port is outside the bandwidth adjustment interval, the convergence port is considered as the bottleneck port.
  • the parameter information includes the traffic and bandwidth of the convergence port.
  • the convergence port is looped one by one, and it is determined whether the bandwidth adjustment value of each convergence port is outside the bandwidth adjustment interval, and if the bandwidth adjustment value of the convergence port is outside the bandwidth adjustment interval, The convergence port is a bottleneck port. If the bandwidth adjustment value of the convergence port is within the bandwidth adjustment interval, the convergence port is a resource-sufficient port. For a resource-sufficient port, at least one of the optimal convergence ratio adjustment value and the utilization adjustment value is directly calculated according to the bandwidth adjustment value and the user-configured policy, so as to optimize the convergence ratio.
  • the port to be adjusted that needs to be optimized for convergence ratio is determined.
  • the working pseudowire and the protection pseudowire correspond to their respective work and protection tunnels.
  • a total of four service tunnels ensure the same bandwidth; among them, the work pseudowire corresponds to two service tunnels, namely the working tunnel and the protection tunnel, and the protection pseudowire Corresponding to two service tunnels, namely a working tunnel and a protection tunnel. Therefore, there are 4 business tunnels. Among them, the pseudowire is built on the business tunnel.
  • the tunnel where the pseudowire is located and the tunnel corresponding to the pseudowire protection group where the pseudowire is located have the same guaranteed bandwidth; that is, the tunnel where the pseudowire is located is a pseudowire tunnel, and the tunnel corresponding to the pseudowire protection group where the pseudowire is located To protect the group tunnel, the guaranteed bandwidth of the pseudowire tunnel is the same as the guaranteed bandwidth of the protection group tunnel.
  • the guaranteed bandwidth of all related work, protection, and protection group tunnels under one service tunnel is the same; that is, the guaranteed bandwidth of the working tunnel, protection tunnel, and protection group tunnel associated with the service tunnel is the same.
  • step S103 when calculating the actual bandwidth adjustment value of the bottleneck port, first find the combination of the service tunnel group under the bottleneck port, and then calculate the actual bandwidth adjustment value of each service tunnel group in the combination.
  • the service tunnel group includes at least one service tunnel, and the packets of the service tunnel are grouped according to the grouping principle described in the step S102 to ensure that the service tunnel bandwidth adjustment of the same group is consistent. Then each service tunnel group is combined to construct a tunnel-port relationship coefficient matrix A of the bottleneck port. By reducing or increasing the bandwidth adjustment value of each service tunnel group within the combination, it satisfies the overall port adjustment requirement on the subnet.
  • the convergence ratio is optimized, the corresponding optimal convergence ratio adjustment value and the utilization adjustment value are calculated, and finally the calculation result is output to the back-end system for further processing. Processing or application.
  • the bandwidth adjustment value of the service tunnel group through the bottleneck port (the bottleneck port is the convergence port) needs to be consistent, as the bandwidth adjustment of the same group of service tunnels is consistent.
  • the other non-converged ports that the service tunnel group passes will also calculate the actual bandwidth adjustment value, that is, the passive adjustment of the bottleneck port.
  • A is the service tunnel-port relationship coefficient matrix under the bottleneck port
  • b is the ideal bandwidth adjustment value of the bottleneck port
  • c m a 1 x 1 + a 2 x 2 +...+a n x n
  • c m is the actual bandwidth adjustment value of the bottleneck port m
  • a n is the bandwidth adjustment value of the service tunnel group under the bottleneck port n, that is, a n is the bottleneck port corresponding to the matrix
  • X ⁇ x 1 , x 2 , ... x n ⁇ , where n is a positive integer.
  • the error analysis condition is: the actual bandwidth adjustment value of the bottleneck port (ie, the convergence port) is adjusted in the bandwidth.
  • the actual bandwidth adjustment value of the non-convergence port is less than or equal to the initial bandwidth adjustment value.
  • the convergence ratio of the bottleneck port is optimized according to the actual bandwidth adjustment value c. If the actual bandwidth adjustment value c does not satisfy the above error analysis condition, the following calculation process is performed:
  • a new service tunnel group is added to the A matrix, and the corresponding b-vectors are also modified synchronously.
  • the combination of the entire tunnel group is calculated again, that is, the bandwidth adjustment value of all the added service tunnel groups is calculated according to the above calculation manner.
  • the third column is fixedly combined: 1, 2, 3, 1, 3, 2, 3;
  • the fourth column is fixedly combined: 1, 4; 2, 4; 3, 4; 1, 2, 4; 1, 3, 4; 2, 3, 4; 1, 2, 3, 4; Until the actual bandwidth adjustment value of all the added service tunnel groups can be obtained, and the error analysis condition is satisfied, and the error is satisfied.
  • the convergence ratio of the bottleneck port is optimized according to the obtained actual bandwidth adjustment value; otherwise, the loop is continued until the 10 groups of service tunnel groups are stopped.
  • the numbers involved in this embodiment such as the construction of two sets of service tunnel groups of the bottleneck port, the addition of one service tunnel group when the error analysis condition is not satisfied, and the addition to the 10 groups of service tunnel groups are used for the implementation.
  • the examples are explained and are not intended to limit the application. Of course, it can also be constructed by using at least two sets of service tunnels of the bottleneck port. When the error analysis condition is not satisfied, at least one service tunnel group can be added, and more than 10 groups of service tunnel groups can be added.
  • a reliable interval is set for b, if the corresponding port in the matrix A is a convergence port, and the bottleneck value is bounded by the bandwidth adjustment interval of the port bandwidth
  • the preset optimal value of the b vector is the upper bound of the bandwidth adjustment interval, and is reduced by 5% to 10%, so that the solution vector X fluctuates around 90% to 95% of the preset optimal value
  • the preset optimal value of the b vector is the lower bound of the bandwidth adjustment interval, and is increased by 5% to 10%, so that the solution vector X is optimal at the preset
  • the value fluctuates from 5% to 10%; if the port is a non-converged port, the preset optimal value of the b vector is directly reduced by 5% to 10%, so that the solution vector X is at a preset optimal value of 90% to
  • the convergence ratio of the bottleneck port is optimized according to the actual bandwidth adjustment value that satisfies the error analysis condition, wherein the convergence ratio adjustment value and the bandwidth utilization adjustment value of the bottleneck port are calculated according to the actual bandwidth adjustment value of the user configuration policy satisfaction and the error analysis condition. At least one of them, and output the calculation result to the back-end system for further processing or application.
  • the user configuration policy is a range of utilization rates or convergence ratios set by the user. For example, the constraint interval for setting the convergence ratio is [0.5, 1]. The smaller the convergence ratio is, the higher the bandwidth utilization is.
  • the convergence ratio is 0.5, according to the relationship between convergence ratio, bandwidth utilization and guaranteed bandwidth. Determine the value of bandwidth utilization.
  • the user sets the utilization range to [0.2, 0.8], and selects the utilization rate as 0.5, and determines the value of the convergence ratio according to the relationship between the convergence ratio, the bandwidth utilization, and the guaranteed bandwidth.
  • the relationship between convergence ratio, bandwidth utilization and guaranteed bandwidth is as follows:
  • the port optimization method provided by the embodiment of the present invention further utilizes the network resource by using the bandwidth optimization of each port to further perform the finer configuration of the convergence ratio.
  • the entire network can still be optimized under the condition of insufficient bottleneck port resources, ensuring the rational allocation of the entire network resources, and achieving the purpose of more effectively multiplexing the local bandwidth of the network. .
  • the service tunnel-port relationship coefficient matrix A of the port to be adjusted is constructed based on the first embodiment. Referring to Table 1, the row and column in the table respectively correspond to the row and column of the matrix, and the data in the table is the element value of the matrix.
  • the port to be adjusted includes a convergence port and a non-convergence port.
  • the table includes four bottleneck ports, which are P1, P2, P3, and P4.
  • Each port has a combination of two service tunnel groups, which are respectively recorded as L1 and L2.
  • the service tunnel group corresponding to L1 is ⁇ L11, L12 ⁇
  • the service tunnel group corresponding to L2 is ⁇ L21, L22 ⁇
  • the port through which the L11 service tunnel group passes is ⁇ P1, P2 ⁇
  • the port through which the L12 service tunnel group passes is ⁇ P2.
  • P3 ⁇ the port through which the L21 service tunnel group passes is ⁇ P1, P2 ⁇
  • the port through which the L22 service tunnel group passes is ⁇ P3, P4 ⁇
  • the P1 port is a convergence port, and the remaining ports are non-convergence ports.
  • the optimal bandwidth adjustment interval of P1 is [10, 45], and the actual service bandwidth adjustment needs 50M.
  • the remaining ports ie, P2, P3, and P4 adjust the bandwidth to 70M, 50M, and 30M, respectively, and the combination L1 of the service tunnel group.
  • the adjustment bandwidth corresponding to L2 is 20M and 30M respectively.
  • each port is 20M; in the combination of L2 service tunnel groups, each port is 30M.
  • the L11 service tunnel group passes through the port ⁇ P1, P2 ⁇
  • the L12 service tunnel group passes through the port ⁇ P2, P3 ⁇ , that is, the L11 and L12 service tunnel groups pass through the port P2.
  • the combination of the L1 service tunnel group has an adjustment bandwidth of 40M on the P2 port.
  • the adjustment bandwidth of the port P4 is 0M.
  • the actual service bandwidth adjustment requires 50M, but the upper bound of the bandwidth adjustment interval is 45M, therefore, the actual service bandwidth is adjusted to the upper bound of the bandwidth adjustment interval, and the initial ideal adjustment output vector b of the port is [45, 70, 50, 30].
  • the bottleneck port bandwidth adjustment vector c satisfies the error analysis condition, that is, the adjustment value of the convergence port P1 is 43.8M in the bandwidth adjustment interval [10, 45], and the adjustment values of the non-convergence ports P2, P3, and P4 are smaller
  • the bandwidth adjustment output vector c is output to the backend system for further processing or application.
  • FIG. 2 is a port optimization apparatus according to the embodiment, where the apparatus includes:
  • the bottleneck port determining module 201 is configured to determine a bottleneck port in the convergence port, where the bottleneck port is a convergence port whose bandwidth adjustment value is outside the bandwidth adjustment interval;
  • the bandwidth calculation module 202 is configured to calculate an actual bandwidth adjustment value of the bottleneck port
  • the convergence ratio optimization module 203 is configured to converge the bottleneck port according to the actual bandwidth adjustment value. Better than optimization.
  • the bottleneck port determining module 201 includes: a bandwidth optimization sub-module 2011, a bandwidth selection sub-module 2012, and a bottleneck port determining sub-module 2013;
  • the bandwidth optimization sub-module 2011 is set to optimize the bandwidth of each port
  • the bandwidth selection sub-module 2012 is configured to select a bandwidth adjustment value of the convergence port from the optimization result
  • the bottleneck port determination sub-module 2013 is configured to determine a bottleneck port in the convergence port according to the bandwidth adjustment value and the bandwidth adjustment interval.
  • the bandwidth calculation module 202 includes: a lookup submodule 2021 and a bandwidth calculation submodule 2022;
  • the lookup submodule 2021 is configured to look up a combination of service tunnel groups under the bottleneck port;
  • the bandwidth calculation sub-module 2022 is configured to calculate an actual bandwidth adjustment value for the combination of the service tunnel groups.
  • the convergence ratio optimization module 203 includes: a determination sub-module 2031, a first convergence ratio optimization sub-module 2032, and a second convergence ratio optimization sub-module 2033;
  • the determining sub-module 2031 is configured to determine whether the actual bandwidth adjustment value satisfies an error analysis condition, where the actual bandwidth adjustment value of the convergence port is within the bandwidth adjustment interval;
  • the first convergence ratio optimization sub-module 2032 is configured to optimize the convergence ratio of the bottleneck port according to the actual bandwidth adjustment value when the actual bandwidth adjustment value satisfies the error analysis condition;
  • the second convergence ratio optimization sub-module 2033 is configured to add at least one service tunnel group when the actual bandwidth adjustment value does not satisfy the error analysis condition, and calculate an actual bandwidth adjustment value of the combination of all the added service tunnel groups until the actual bandwidth is obtained.
  • the adjustment value satisfies the error analysis condition.
  • the bottleneck port determining module 201 first collects bandwidth, traffic, convergence ratio assets and performance data of all ports (including convergence ports and non-convergence ports) of a certain subnet resource in a certain period during bandwidth optimization.
  • the parameter information is then optimized according to the selected parameter information and the bandwidth optimization strategy for each port bandwidth in the period to solve the problem that the port bandwidth is insufficient and the bandwidth adjustment cannot be performed.
  • the bandwidth optimization strategy is to provide an initial bandwidth adjustment value of the service tunnel under the port and an initial bandwidth adjustment value of the pseudowire carried by the service tunnel according to the collected parameter information. When optimizing the bandwidth of each port, the optimization method can be used.
  • the optimization method such as dynamically adjusting the bandwidth currently configured by the port according to the bandwidth required by the user to meet the user's needs, solves the problem of insufficient port bandwidth.
  • the bandwidth adjustment value of each port and the service tunnel bandwidth adjustment value in each port are obtained.
  • the port includes a convergence port and a non-convergence port, and then selects a bandwidth adjustment value of at least one convergence port from the port. Select the bandwidth adjustment value of all converged ports from the port.
  • the bandwidth adjustment interval of the convergence port is determined by the parameter information of the convergence port, and then the bandwidth adjustment value of the convergence port is determined to be outside the bandwidth adjustment interval. If the bandwidth adjustment value of the convergence port is outside the bandwidth adjustment interval, The convergence port is considered to be a bottleneck port.
  • the parameter information includes the traffic and bandwidth of the convergence port.
  • the bandwidth adjustment interval is determined by the bandwidth utilization and the port convergence ratio after the bandwidth optimization and the parameter information of the parameter.
  • the bandwidth utilization can be manually configured, and the convergence ratio can also be manually configured through policies.
  • the bandwidth adjustment value of the selected convergence port is within the bandwidth adjustment interval. If not within the bandwidth adjustment interval, the convergence port is a bottleneck port.
  • the convergence port is looped one by one, and it is determined whether the bandwidth adjustment value of each convergence port is outside the bandwidth adjustment interval, and if the bandwidth adjustment value of the convergence port is outside the bandwidth adjustment interval, The convergence port is a bottleneck port. If the bandwidth adjustment value of the convergence port is within the bandwidth adjustment interval, the convergence port is a resource-sufficient port. For a resource-sufficient port, at least one of the optimal convergence ratio adjustment value and the utilization adjustment value is directly calculated according to the bandwidth adjustment value and the user-configured policy, so as to optimize the convergence ratio.
  • the service tunnel group includes at least one service tunnel, and the packets of the service tunnel are grouped according to the grouping principle described in the first embodiment to ensure that the bandwidth adjustment of the service tunnels of the same group is consistent. Then each service tunnel group is combined to construct a tunnel-port relationship coefficient matrix A of the bottleneck port. By reducing or increasing the number of each service tunnel group within the combination The bandwidth adjustment value makes it meet the overall port adjustment requirements on the subnet.
  • the convergence ratio is optimized, the corresponding optimal convergence ratio adjustment value and the utilization adjustment value are calculated, and finally the calculation result is output to the back-end system for further processing. Processing or application.
  • the bandwidth adjustment value of the service tunnel group through the bottleneck port (the bottleneck port is the convergence port) needs to be consistent, as the bandwidth adjustment of the same group of service tunnels is consistent.
  • the other non-converged ports that the service tunnel group passes will also calculate the actual bandwidth adjustment value, that is, the passive adjustment of the bottleneck port.
  • the bandwidth ratio optimization of each port is further performed to further fine-tune the convergence ratio, thereby more effectively utilizing network resources.
  • the entire network can still be optimized under the condition of insufficient bottleneck port resources, ensuring the rational allocation of the entire network resources, and achieving the purpose of more effectively multiplexing the local bandwidth of the network. .
  • the embodiment further provides a port optimization apparatus, the apparatus includes a processor, and the processor is configured to implement the functions of the bottleneck port determining module 201, the bandwidth calculating module 202, and the convergence ratio optimization module 203, and perform the following process: determining convergence A bottleneck port in the port is a convergence port whose bandwidth adjustment value is outside the bandwidth adjustment interval. The actual bandwidth adjustment value of the bottleneck port is calculated, and the convergence ratio of the bottleneck port is optimized according to the actual bandwidth adjustment value.
  • the processor can be arranged to implement the functions performed by each of the sub-modules described above.
  • the embodiment also provides an optimization system comprising a port and the above port optimization device.
  • modules or steps of the embodiments of the present invention may be implemented by a general-purpose computing device, which may be centralized on a single computing device or distributed over a network of multiple computing devices.
  • the program code executable by the computing device can be implemented, so that they can be stored in a storage medium (ROM/RAM, disk, optical disk) by the computing device, and in some cases, can be different from this
  • ROM/RAM, disk, optical disk a storage medium
  • the steps shown or described are performed sequentially, or they are separately fabricated into integrated circuit modules, or a plurality of modules or steps thereof are fabricated as a single integrated circuit module. Therefore, the embodiment of the present invention does not Limited to any specific hardware and software combination.
  • the bandwidth of the bottleneck port is adjusted by the joint optimization of the port bandwidth and the convergence ratio, so as to ensure that the port convergence ratio, the guaranteed bandwidth, and the port utilization are dynamically balanced under the condition of sufficient resources, so that The port bandwidth is finerly optimized.
  • the joint optimization method further optimizes the convergence ratio based on the bandwidth optimization of all ports in the optimized range to more effectively utilize network resources.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本文公布一种端口优化方法、装置以及系统，确定收敛端口中的瓶颈端口，该瓶颈端口为带宽调整值在带宽调整区间外的收敛端口，计算瓶颈端口的实际带宽调整值，并根据实际带宽调整值对所述瓶颈端口进行收敛比优化。

Description

一种端口优化方法、装置及系统 技术领域
本文涉及但不限于电信和企业网络管理领域，尤指一种端口优化方法、装置及系统。
背景技术
在网络端口带宽进行初步优化后端口带宽利用逐渐趋于合理，但是由于潮汐效应等因素，网络端口中所有业务实际流速不会同时达到业务的保证带宽。为了更好的利用已建设的网络资源，通过对网络端口的带宽进行优化，以实现网络资源的合理利用。但是传统的带宽优化方式只能针对大部分端口带宽进行粗略优化，无法对端口带宽做出更精细的优化，不能充分的利用网络资源，从而不能保证更有效的复用网络局部带宽，进而影响用户使用网络时的体验。
发明概述
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
本发明实施例提供一种端口优化方法、装置及系统，以充分的利用网络资源。
本发明实施例提供一种端口优化方法，包括：
确定收敛端口中的瓶颈端口，所述瓶颈端口为带宽调整值在带宽调整区间外的收敛端口；
计算所述瓶颈端口的实际带宽调整值；
根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化。
在本发明一种实施方式中，所述确定收敛端口中的瓶颈端口包括：
对每个端口的带宽进行优化；
从优化结果中选择收敛端口的带宽调整值；
根据带宽调整值和带宽调整区间确定收敛端口中的瓶颈端口。
在本发明一种实施方式中，计算所述瓶颈端口的实际带宽调整值包括：
查找所述瓶颈端口下的业务隧道组的组合；
计算所述业务隧道组的组合的实际带宽调整值。
在本发明一种实施方式中，根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化包括：
判断所述实际带宽调整值是否满足误差分析条件，所述误差分析条件为瓶颈端口的实际带宽调整值在带宽调整区间内；
当所述实际带宽调整值满足误差分析条件时，根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化；
当所述实际带宽调整值不满足误差分析条件时，新增至少一个业务隧道组，计算增加后的所有业务隧道组的组合的实际带宽调整值，直到该实际带宽调整值满足误差分析条件。
在本发明一种实施方式中，根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化包括：
根据用户配置策略和满足误差分析条件的实际带宽调整值计算所述瓶颈端口的收敛比调整值和带宽利用率调整值中的至少一个，并输出计算结果。
在本发明一种实施方式中，计算所述瓶颈端口的实际带宽调整值包括：
根据超定方程AX＝b确定X，并根据cm＝a1x1+a2x2+…+anxn确定cm，X为带宽调整解向量，A为瓶颈端口下的业务隧道-端口关系系数矩阵，b为瓶颈端口的理想带宽调整值；cm为瓶颈端口m的实际带宽调整值，an为矩阵A中该瓶颈端口对应的元素，X＝{x1，x2，…xn}，n和m均为正整数。
本发明实施例还提供了一种端口优化装置，包括：
瓶颈端口确定模块，设置为确定收敛端口中的瓶颈端口，所述瓶颈端口为带宽调整值在带宽调整区间外的收敛端口；
带宽计算模块，设置为计算所述瓶颈端口的实际带宽调整值；
收敛比优化模块，设置为根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化。
在本发明一种实施方式中，所述瓶颈端口确定模块包括：
带宽优化子模块，设置为对每个端口的带宽进行优化；
带宽选择子模块，设置为从优化结果中选择收敛端口的带宽调整值；
瓶颈端口确定子模块，设置为根据带宽调整值和带宽调整区间确定收敛端口中的瓶颈端口。
在本发明一种实施方式中，所述带宽计算模块包括：
查找子模块，设置为查找所述瓶颈端口下的业务隧道组的组合；
带宽计算子模块，设置为计算所述业务隧道组的组合的实际带宽调整值。
在本发明一种实施方式中，所述收敛比优化模块包括：
判断子模块，设置为判断所述实际带宽调整值是否满足误差分析条件，所述误差分析条件为收敛端口的实际带宽调整值在收敛区间内；
第一收敛比优化子模块，设置为当所述实际带宽调整值满足误差分析条件时，根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化；
第二收敛比优化子模块，设置为当所述实际带宽调整值不满足误差分析条件时，新增至少一个业务隧道组，计算增加后的所有业务隧道组的组合的实际带宽调整值，直到该实际带宽调整值满足误差分析条件。
本发明实施例还提供了一种端口优化系统，包括：端口以及如上所述的端口优化装置。
本发明实施例还提供了一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行前述的任一项的端口优化方法。
本发明实施例提供了一种端口优化方法、装置、系统以及计算机存储介质，通过确定收敛端口中的瓶颈端口，该瓶颈端口为带宽调整值在带宽调整区间外的收敛端口，计算瓶颈端口的实际带宽调整值，并根据实际带宽调整值对所述瓶颈端口进行收敛比优化。本发明实施例通过端口带宽与收敛比联合优化的方式，对瓶颈端口的带宽进行调整，以保证端口收敛比、保证带宽、端口利用率三者之间在资源充足的条件可以保持动态平衡，使得端口带宽做出更精细的优化，其中，联合优化方法即为对待优化范围内所有端口进行带宽优化的基础上进一步进行收敛比的精细化配置，以更有效的利用网络资源。
在阅读并理解了附图和详细描述后，可以明白其他方面。
附图概述
图1为本发明实施例一提供的端口优化方法流程图；
图2为本发明实施例三提供的端口优化装置示意图。
详述
下面通过实施方式结合附图对本发明实施例作进一步详细说明。
实施例一
本实施例提供一种端口优化方法，请参见图1，其步骤如下：
S101，确定收敛端口中的瓶颈端口，该瓶颈端口为带宽调整值在带宽调整区间外的收敛端口。
系统采集某子网资源在某个周期内所有端口(包括收敛端口和非收敛端口)的带宽、流量、收敛比的资产和性能数据等参数信息，然后根据选定的参数信息以及带宽优化策略对该周期内的每个端口带宽进行优化。其中，带宽优化策略为根据采集的参数信息给出端口下业务隧道的初始带宽调整值，以及业务隧道承载的伪线的初始带宽调整值。带宽优化完成后，从优化结果中选择收敛端口的带宽调整值，并根据带宽调整值和带宽调整区间确定收敛端口中的瓶颈端口。
其中，带宽调整区间是通过带宽优化后的带宽利用率和端口收敛比两个区间确定的区间。其中，带宽利用率可以进行人为配置，收敛比也可以通过策略进行人为配置。带宽调整区间是根据采集的参数信息以及收敛比优化策略设定的一个区间。其中，收敛比优化策略为根据获取的参数信息寻求合理的端口收敛比以及带宽利用率。另一实施方式中，带宽调整区间也可由用户根据采集的参数信息进行设定。
此外，前述瓶颈端口是指端口网络资源不足，在保证端口利用率的前提下无法通过调整端口收敛比提供足够的带宽来配置相应的业务需求。收敛端口中包括至少一个瓶颈端口，若收敛端口中不包括瓶颈端口，则认为进行带宽优化的每个端口资源充足，可直接进行收敛比优化，无需对各自的带宽调整值进行调整。
S102，计算瓶颈端口的实际带宽调整值。
其中，利用超定方程AX＝b的最小二乘近似解寻找瓶颈端口下的业务隧道组的组合，然后计算每个业务隧道组的实际带宽调整值，从而确定瓶颈端口的实际带宽调整值，实现对瓶颈端口的收敛比优化。此外，对瓶颈端口端口进行收敛比优化的前提是端口保证带宽调整值、端口收敛比和带宽利用率三者平衡。
S103，根据实际带宽调整值对瓶颈端口进行收敛比优化。
其中，对瓶颈端口端口进行收敛比优化的前提是端口保证带宽调整值、端口收敛比和带宽利用率三者平衡。通过计算收敛比调整值和带宽利用率调整值中的至少一个实现收敛比的优化。
在S101步骤中，对每个端口带宽的优化方式可采用已有的优化方式，如根据用户所需的带宽动态调整端口当前配置的带宽以满足用户需求，解决端口带宽剩余不足的问题。带宽优化完毕后，得到每个端口的带宽调整值，以及每个端口中的业务隧道带宽调整值，端口包括收敛端口和非收敛端口，然后从端口中选择至少一个收敛端口的带宽调整值，可以从端口中选择所有收敛端口的带宽调整值。
在确定收敛端口中的瓶颈端口时，通过采集收敛端口的参数信息，确 定收敛端口的带宽调整区间，然后判断收敛端口的带宽调整值是否处于带宽调整区间外，若收敛端口的带宽调整值处于带宽调整区间外，则认为该收敛端口为瓶颈端口。其中，参数信息包括收敛端口的流量和带宽。
其中，当端口中包括至少两个收敛端口时，则逐个循环收敛端口，判断每个收敛端口的带宽调整值是否在带宽调整区间外，若收敛端口的带宽调整值在带宽调整区间外，则该收敛端口为瓶颈端口，若收敛端口的带宽调整值在带宽调整区间内，则该收敛端口为资源充足端口。对于资源充足端口，直接根据其带宽调整值和用户配置的策略计算对应的最佳收敛比调整值和利用率调整值中的至少一个，实现收敛比的优化。对于瓶颈端口，利用超定方程AX＝b的最小二乘近似解计算瓶颈端口的实际带宽调整值，根据该实际带宽调整值和用户配置的策略计算对应的最佳收敛比调整值和利用率调整值中的至少一个。
其中，端口进行宽优化后，确定需要进行收敛比优化的待调整端口。假设待调整的业务隧道记为L＝{L1，L2，L3，…Ln}，共计n条业务隧道；它们经过的全部端口记为P＝{P1，P2，P3，…Pi}，共计i个端口，这i个端口即为待调整端口(包括收敛端口和非收敛端口)。若i个待调整端口中有k个待调整端口下的业务资源分配充足，则还剩余(i-k)个待调整端口下的资源分配不足。
将L条待调整业务隧道分为H组，分组后业务隧道记为Lh＝{Lh1，Lh2，…Lh(rh)}，其中，rh为大于等于1的正整数。在对业务隧道进行分组时，要保证同一组业务隧道的带宽调整保持一致。业务隧道分组时遵循以下四个原则：
1)共享隧道的保证带宽相同；
2)工作伪线与保护伪线分别对应各自的工作、保护隧道，共4条业务隧道保证带宽相同；其中，工作伪线对应两条业务隧道，分别是工作隧道和保护隧道，保护伪线也对应两条业务隧道，分别是工作隧道和保护隧道。因此，共有4条业务隧道。其中，伪线建立在业务隧道之上。
3)伪线所在的隧道与伪线所在的伪线保护组对应的隧道的保证带宽相同；也即伪线所在的隧道为伪线隧道，伪线所在的伪线保护组对应的隧道 为保护组隧道，伪线隧道的保证带宽与保护组隧道的保证带宽相同。
4)一条业务隧道下所有相关的工作、保护以及保护组隧道的保证带宽相同；即与业务隧道相关的工作隧道、保护隧道以及保护组隧道三者的保证带宽相同。
待调整业务隧道分组后，将每条业务隧道经过的端口构成集合，找到业务隧道组Lh所对应的端口组记为LPh＝{LPh1，LPh2，LPh3，…LPh(th)}；其中，th为大于等于1的正整数。寻找待调整端口Pi下经过的全部业务隧道组，并构成集合PLi＝{PLi1，PLi2，PLi3，…PLi(si)}，其中，i和si均为大于等于1的正整数。逐个循环待调整端口，判断该待调整端口的带宽调整值是否在带宽调整区间内，若不在带宽调整区间内，则该待调整端口为瓶颈端口，反之为资源充足端口。
在S103步骤，在对瓶颈端口的实际带宽调整值进行计算时，首先查找瓶颈端口下的业务隧道组的组合，然后计算组合中每个业务隧道组的实际带宽调整值。业务隧道组中包括至少一个业务隧道，且业务隧道的分组按照S102步骤中描述的分组原则进行分组，以保证同一组的业务隧道带宽调整保持一致。然后每个业务隧道组进行组合，构造出瓶颈端口的隧道-端口关系系数矩阵A。通过降低或提高该组合内的每个业务隧道组的带宽调整值使其满足子网上的整体端口调整需求。然后，根据计算的每个端口的实际带宽调整值，以及用户配置的策略进行收敛比优化，计算对应的最佳收敛比调整值、利用率调整值，最后将计算结果输出到后端系统进一步加工处理或应用。
在计算瓶颈端口的实际带宽调整值的同时，由于同一组业务隧道的带宽调整保持一致，因此，经过该瓶颈端口(该瓶颈端口为收敛端口)的业务隧道组的带宽调整值需要保持一致，相应的，业务隧道组经过的其他非收敛端口也会计算实际带宽调整值，也即被动的跟随瓶颈端口的调整而调整。
首先，计算瓶颈端口的实际带宽调整值时，其过程如下：
构造端口Pi的业务隧道-端口关系系数矩阵A，该矩阵列标对应的是该端口所包含的业务隧道组即PLi，行标为每组业务隧道所经过的全部端口带 宽调整值，由于每组业务隧道所经过的全部端口带宽调整值是一样的，所以矩阵A的每个元素值为某一隧道组的期望带宽调整值。对于同一组隧道中有s条隧道经过了同一端口，那么对应的行列所在的元素值为调整值的s倍。秉着对网络影响尽可能小的变动原则，本实施例以一个瓶颈端口的2组业务隧道组开始构造，即最先取两组隧道，整理所有涉及行标中所有端口的带宽调整值(即端口Pi的理想带宽调整值，该理想带宽调整值为经过该端口的每个业务隧道组的理想带宽调整值之和)，用来构造最佳端口调整值b。然后，利用超定方程AX＝b的最小二乘近似解，求得带宽调整解向量X，根据带宽调整解向量X计算出每个瓶颈端口的实际带宽调整值c，实际带宽调整值c包括每个端口的实际带宽调整值，也即c＝{c1，c2，…cm}，n和m均大于0，且为正整数，该端口包括瓶颈端口和不收敛端口。
其中，根据超定方程AX＝b确定X，A为瓶颈端口下的业务隧道-端口关系系数矩阵，b为瓶颈端口的理想带宽调整值；然后根据cm＝a1x1+a2x2+…+anxn确定cm，cm为瓶颈端口m的实际带宽调整值，an为瓶颈端口n下的业务隧道组的带宽调整值，即an为矩阵A中该瓶颈端口对应的元素，X＝{x1，x2，…xn}，n为正整数。
然后，计算出瓶颈端口的实际带宽调整值c后，对c进行误差分析，判断其是否满足误差分析条件，其中，误差分析条件为：瓶颈端口(即收敛端口)的实际带宽调整值在带宽调整区间内，无收敛端口的实际带宽调整值小于等于最初带宽调整值。在满足上述误差分析条件的情况下，根据实际带宽调整值c对瓶颈端口进行收敛比优化。若实际带宽调整值c不满足上述误差分析条件，则执行如下计算过程：
在A矩阵中新增一个业务隧道组，相应的b向量也同步修改，重新对整个隧道组的组合进行计算，也即按照上述计算方式计算增加后的所有业务隧道组的带宽调整值。计算时每次以新加入的最后一个业务隧道组为定点，进行排列组合，如3列业务隧道组时，第三列固定进行组合：1、2、3，1、3，2、3；4列时，第四列固定进行组合：1、4；2、4；3、4；1、2、4；1、3、4；2、3、4；1、2、3、4；依次类推，直到可以得出增加后的所有业务隧道组的实际带宽调整值，并且满足误差分析条件，在满足误 差分析条件的情况下，根据得到的实际带宽调整值对瓶颈端口进行收敛比优化；否则继续循环，直到增加到10组业务隧道组停止。本实施例中涉及到的数字，如以瓶颈端口的2组业务隧道组进行构造、不满足误差分析条件时增加一个业务隧道组、直到增加到10组业务隧道组等表述均是用于对本实施例进行解释，并非用于限定申请。当然也可以以瓶颈端口的至少两组业务隧道进行构造，不满足误差分析条件时也可增加至少一个业务隧道组，以及增加到多于10组业务隧道组等均可。
在实际应用中，为了提升解向量X的误差分析概率，可以使用以下方案：对b设置一个可靠区间，若矩阵A中所对应端口为收敛端口，且瓶颈值在端口带宽调整带宽调整区间上界以上，则b向量的预设最佳值为带宽调整区间上界，并将其缩小5％～10％，使得解向量X在预设最佳值90％～95％左右波动；若该端口为收敛端口，且瓶颈值在端口带宽调整带宽调整区间下界以下，则b向量的预设最佳值为带宽调整区间下界，并将其增加5％～10％，使得解向量X在预设最佳值5％～10％左右波动；若该端口为无收敛端口，则直接将b向量的预设最佳值缩小5％～10％，使得解向量X在预设最佳值90％～95％左右波动；经过上述调整过程，则认为调整后的带宽均是可靠的。对于调整后依然不满足误差分析条件的端口以及端口下的业务隧道，则进入舍弃策略，即放弃端口下的某些次要业务，从而降低端口调整需求，或者给出用户提示需要对网络扩容。
最后，根据满足误差分析条件的实际带宽调整值对瓶颈端口进行收敛比优化，其中，根据用户配置策略满足和误差分析条件的实际带宽调整值计算瓶颈端口的收敛比调整值和带宽利用率调整值中的至少一个，并将计算结果输出到后端系统进一步加工处理或应用。其中，用户配置策略为，用户设定的利用率范围或收敛比范围。例如：用户设定收敛比的约束区间为[0.5，1]，由于收敛比越小，带宽利用率越高，则取收敛比为0.5，根据收敛比、带宽利用率和保证带宽之间的关系，确定带宽利用率的取值。或者，用户设定利用率范围为[0.2，0.8]，选取利用率为0.5，则根据收敛比、带宽利用率和保证带宽之间的关系确定收敛比的取值。其中，收敛比、带宽利用率以及保证带宽之间的关系如下：
Figure PCTCN2017078287-appb-000001
本发明实施例提供的端口优化方法，通过在对每个端口进行带宽优化的基础上进一步进行收敛比的精细化配置，更有效的利用网络资源。同时，通过对瓶颈端口下的业务隧道组的带宽优化，在瓶颈端口资源不足的条件下依然能够进行全网优化，保证了全网资源的合理分配，达到更有效的复用网络局部带宽的目的。
实施例二
本实施将以实施例一为基础，构造待调整端口的业务隧道-端口关系系数矩阵A，请参见表1，表中行和列分别对应矩阵的行列，表中数据即为矩阵的元素值。其中，待调整端口包括收敛端口和非收敛端口。
表1
Figure PCTCN2017078287-appb-000002
针对表1的分析，过程如下：
表中包括4个瓶颈端口，分别是P1，P2，P3，P4，每个端口下有2个业务隧道组的组合，分别记为L1，L2。L1对应的业务隧道组为{L11，L12}，L2对应的业务隧道组为{L21，L22}，L11业务隧道组经过的端口为{P1，P2}，L12业务隧道组经过的端口为{P2，P3}，L21业务隧道组经过的端口为{P1，P2}，L22业务隧道组经过的端口为{P3，P4}，其中P1端口为收敛端口，其余端口均为无收敛端口。假设P1最佳带宽调整区间为[10，45]，而实际业务带宽调整需要50M，其余端口(即P2，P3和P4)调整带宽分别为70M，50M，30M，而业务隧道组的组合L1，L2对应的调整带宽分别为20M，30M。其中，在对业务隧道进行分组时，在满足分组原则的 情况下，也需要同时满足带宽调整值一致。因此，在L1业务隧道组的组合中，每个端口均为20M；在L2业务隧道组的组合中，每个端口均为30M。但是，在L1业务隧道组的组合中，L11业务隧道组经过端口{P1，P2}，L12业务隧道组经过端口{P2，P3}，也即L11与L12两个业务隧道组共同经过端口P2，则L1业务隧道组的组合在P2端口的调整带宽为40M。此外，由于L1业务隧道组的组合均为经过端口P4，则端口P4的调整带宽为0M。
基于上述条件，因为实际业务带宽调整需要50M，但是带宽调整区间上界为45M，因此，将实际业务带宽调整成带宽调整区间上界，则端口的初始理想调整输出向量b为[45，70，50，30]。将每个端口的初始理想调整输出向量缩小10％，则输出的带宽调整输出向量b为[40.5，63，45，27]；对超定方程AX＝b进行求解，则对应的解向量X约为[0.90，0.86]；将解向量X代入cm＝a1x1+a2x2+…+anxn，则最终输出的瓶颈端口带宽调整向量c约为[43.8，61.8，43.8，25.8]；将求得的解向量X分别对应着与L1和L2的调整带宽相乘，得到L1和L2两个业务隧道组的组合的最终带宽调整值约为[18.00，25.8]；上述输出的瓶颈端口带宽调整向量c满足误差分析条件，即收敛端口P1的调整值43.8M落在带宽调整区间[10，45]内，无收敛端口P2、P3和P4的调整值均小于最初的带宽调整值70M，50M，30M。最后将带宽调整输出向量c输出至后端系统进一步加工处理或应用。通过本实施例构造的矩阵A，保证了全网资源的合理分配，能够在瓶颈端口不足条件下进行全网优化，从而达到更有效的复用网络局部带宽的目的。
实施例三
上述实施例一中的端口优化方法可应用于本实施例提供的端口优化装置，故本实施例不再对装置中的部分模块做详细说明，内容请参见实施例一中的相关描述。请参见图2，图2为本实施例提供的端口优化装置，该装置包括：
瓶颈端口确定模块201，设置为确定收敛端口中的瓶颈端口，瓶颈端口为带宽调整值在带宽调整区间外的收敛端口；
带宽计算模块202，设置为计算所述瓶颈端口的实际带宽调整值；
收敛比优化模块203，设置为根据实际带宽调整值对瓶颈端口进行收敛 比优化。
瓶颈端口确定模块201包括：带宽优化子模块2011、带宽选择子模块2012以及瓶颈端口确定子模块2013；
带宽优化子模块2011设置为对每个端口的带宽进行优化；
带宽选择子模块2012设置为从优化结果中选择收敛端口的带宽调整值；
瓶颈端口确定子模块2013设置为根据带宽调整值和带宽调整区间确定收敛端口中的瓶颈端口。
带宽计算模块202包括：查找子模块2021和带宽计算子模块2022；
查找子模块2021设置为查找瓶颈端口下的业务隧道组的组合；
带宽计算子模块2022设置为计算业务隧道组的组合的实际带宽调整值。
收敛比优化模块203包括：判断子模块2031、第一收敛比优化子模块2032以及第二收敛比优化子模块2033；
判断子模块2031设置为判断实际带宽调整值是否满足误差分析条件，该误差分析条件为收敛端口的实际带宽调整值在带宽调整区间内；
第一收敛比优化子模块2032设置为当实际带宽调整值满足误差分析条件时，根据实际带宽调整值对瓶颈端口进行收敛比优化；
第二收敛比优化子模块2033设置为当实际带宽调整值不满足误差分析条件时，新增至少一个业务隧道组，计算增加后的所有业务隧道组的组合的实际带宽调整值，直到该实际带宽调整值满足误差分析条件。
其中，瓶颈端口确定模块201中，在进行带宽优化时，首先采集某子网资源在某个周期内所有端口(包括收敛端口和非收敛端口)的带宽、流量、收敛比的资产和性能数据等参数信息，然后根据选定的参数信息以及带宽优化策略对该周期内的每个端口带宽进行优化，以解决端口带宽剩余不足，无法进行带宽调整的问题。其中，带宽优化策略为根据采集的参数信息给出端口下业务隧道的初始带宽调整值，以及业务隧道承载的伪线的初始带宽调整值。在对每个端口带宽进行优化时，优化方式可采用已有的 优化方式，如根据用户所需的带宽动态调整端口当前配置的带宽以满足用户需求，解决端口带宽剩余不足的问题。带宽优化完毕后，得到每个端口的带宽调整值，以及每个端口中的业务隧道带宽调整值，端口包括收敛端口和非收敛端口，然后从端口中选择至少一个收敛端口的带宽调整值，可以从端口中选择所有收敛端口的带宽调整值。
带宽优化完毕，通过采集的收敛端口的参数信息，确定收敛端口的带宽调整区间，然后判断收敛端口的带宽调整值是否处于带宽调整区间外，若收敛端口的带宽调整值处于带宽调整区间外，则认为该收敛端口为瓶颈端口。其中，参数信息包括收敛端口的流量和带宽。其中，通过带宽优化后的带宽利用率和端口收敛比两个区间以及参数的参数信息确定带宽调整区间。其中，带宽利用率可以进行人为配置，收敛比也可以通过策略进行人为配置。
确定带宽调整区间后，判断选择的收敛端口的带宽调整值是否在该带宽调整区间内，若不处于该带宽调整区间内，则该收敛端口为瓶颈端口。
其中，当端口中包括至少两个收敛端口时，则逐个循环收敛端口，判断每个收敛端口的带宽调整值是否在带宽调整区间外，若收敛端口的带宽调整值在带宽调整区间外，则该收敛端口为瓶颈端口，若收敛端口的带宽调整值在带宽调整区间内，则该收敛端口为资源充足端口。对于资源充足端口，直接根据其带宽调整值和用户配置的策略计算对应的最佳收敛比调整值和利用率调整值中的至少一个，实现收敛比的优化。对于瓶颈端口，利用超定方程AX＝b的最小二乘近似解计算瓶颈端口的实际带宽调整值，根据该实际带宽调整值和用户配置的策略计算对应的最佳收敛比调整值和利用率调整值中的至少一个。
确定瓶颈端口后，计算瓶颈端口的实际带宽调整值，在进行计算时，首先查找瓶颈端口下的业务隧道组的组合，然后计算组合中每个业务隧道组的实际带宽调整值。业务隧道组中包括至少一个业务隧道，且业务隧道的分组按照实施例一中描述的分组原则进行分组，以保证同一组的业务隧道带宽调整保持一致。然后每个业务隧道组进行组合，构造出瓶颈端口的隧道-端口关系系数矩阵A。通过降低或提高该组合内的每个业务隧道组的 带宽调整值使其满足子网上的整体端口调整需求。然后，根据计算的每个端口的实际带宽调整值，以及用户配置的策略进行收敛比优化，计算对应的最佳收敛比调整值、利用率调整值，最后将计算结果输出到后端系统进一步加工处理或应用。
在计算瓶颈端口的实际带宽调整值的同时，由于同一组业务隧道的带宽调整保持一致，因此，经过该瓶颈端口(该瓶颈端口为收敛端口)的业务隧道组的带宽调整值需要保持一致，相应的，业务隧道组经过的其他非收敛端口也会计算实际带宽调整值，也即被动的跟随瓶颈端口的调整而调整。
通过对上述装置的实施，在对每个端口进行带宽优化的基础上进一步进行收敛比的精细化配置，从而更有效的利用网络资源。同时，通过对瓶颈端口下的业务隧道组的带宽优化，在瓶颈端口资源不足的条件下依然能够进行全网优化，保证了全网资源的合理分配，达到更有效的复用网络局部带宽的目的。
本实施例还提供了一种端口优化装置，该装置包括处理器，处理器设置为实现上述瓶颈端口确定模块201、带宽计算模块202以及收敛比优化模块203的功能，其执行如下过程：确定收敛端口中的瓶颈端口，该瓶颈端口为带宽调整值在带宽调整区间外的收敛端口；计算瓶颈端口的实际带宽调整值，根据实际带宽调整值对所述瓶颈端口进行收敛比优化。此外，处理器还可设置为实现上述每个子模块所执行的功能。
最后，本实施例还提供了一种优化系统，该系统包括端口和上述端口优化装置。
本领域的技术人员应该明白，上述本发明实施例的模块或步骤可以用通用的计算装置来实现，它们可以集中在单个的计算装置上，或者分布在多个计算装置所组成的网络上，它们可以用计算装置可执行的程序代码来实现，从而，可以将它们存储在存储介质(ROM/RAM、磁碟、光盘)中由计算装置来执行，并且在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤，或者将它们分别制作成集成电路模块，或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。所以，本发明实施例不 限制于任何特定的硬件和软件结合。
以上内容是结合实施方式对本发明实施例所作的进一步详细说明，不能认定本申请的实施只局限于这些说明。对于本申请所属技术领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干简单推演或替换，都应当视为属于本申请的保护范围。
工业实用性
本发明实施例通过端口带宽与收敛比联合优化的方式，对瓶颈端口的带宽进行调整，以保证端口收敛比、保证带宽、端口利用率三者之间在资源充足的条件可以保持动态平衡，使得端口带宽做出更精细的优化，其中，联合优化方法即为对待优化范围内所有端口进行带宽优化的基础上进一步进行收敛比的精细化配置，以更有效的利用网络资源。

Claims (11)

  1. 一种端口优化方法，包括：
    确定收敛端口中的瓶颈端口，所述瓶颈端口为带宽调整值在带宽调整区间外的收敛端口；
    计算所述瓶颈端口的实际带宽调整值；
    根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化。
  2. 如权利要求1所述的端口优化方法，其中，所述确定收敛端口中的瓶颈端口包括：
    对每个端口的带宽进行优化；
    从优化结果中选择收敛端口的带宽调整值；
    根据带宽调整值和带宽调整区间确定收敛端口中的瓶颈端口。
  3. 如权利要求1所述的端口优化方法，其中，所述计算所述瓶颈端口的实际带宽调整值包括：
    查找所述瓶颈端口下的业务隧道组的组合；
    计算所述业务隧道组的组合的实际带宽调整值。
  4. 如权利要求1-3任一项所述的端口优化方法，其中，所述根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化包括：
    判断所述实际带宽调整值是否满足误差分析条件，所述误差分析条件为瓶颈端口的实际带宽调整值在带宽调整区间内；
    当所述实际带宽调整值满足误差分析条件时，根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化；
    当所述实际带宽调整值不满足误差分析条件时，新增至少一个业务隧道组，计算增加后的所有业务隧道组的组合的实际带宽调整值，直到该实际带宽调整值满足误差分析条件。
  5. 如权利要求4所述的端口优化方法，其中，根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化包括：
    根据用户配置策略和满足误差分析条件的实际带宽调整值计算所述瓶颈端口的收敛比调整值和带宽利用率调整值中的至少一个，并输出计算结果。
  6. 如权利要求1-3任一项所述的端口优化方法，其中，所述计算所述瓶颈端口的实际带宽调整值包括：
    根据超定方程AX＝b确定X，并根据cm＝a1x1+a2x2+…+anxn确定cm，X为带宽调整解向量，A为瓶颈端口下的业务隧道-端口关系系数矩阵，b为瓶颈端口的理想带宽调整值；cm为瓶颈端口m的实际带宽调整值，an为矩阵A中该瓶颈端口对应的元素，X＝{x1，x2，…xn}，n和m均为正整数。
  7. 一种端口优化装置，包括：
    瓶颈端口确定模块，设置为确定收敛端口中的瓶颈端口，所述瓶颈端口为带宽调整值在带宽调整区间外的收敛端口；
    带宽计算模块，设置为计算所述瓶颈端口的实际带宽调整值；
    收敛比优化模块，设置为根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化。
  8. 如权利要求7所述的端口优化装置，其中，所述瓶颈端口确定模块包括：
    带宽优化子模块，设置为对每个端口的带宽进行优化；
    带宽选择子模块，设置为从优化结果中选择收敛端口的带宽调整值；
    瓶颈端口确定子模块，设置为根据带宽调整值和带宽调整区间确定收敛端口中的瓶颈端口。
  9. 如权利要求7所述的端口优化装置，其中，所述带宽计算模块包括：
    查找子模块，设置为查找所述瓶颈端口下的业务隧道组的组合；
    带宽计算子模块，设置为计算所述业务隧道组的组合的实际带宽调整值。
  10. 如权利要求7-9任一项所述的端口优化装置，其中，所述收敛比优化模块包括：
    判断子模块，设置为判断所述实际带宽调整值是否满足误差分析条件，所述误差分析条件为收敛端口的实际带宽调整值在收敛区间内；
    第一收敛比优化子模块，设置为当所述实际带宽调整值满足误差分析条件时，根据所述实际带宽调整值对所述瓶颈端口进行收敛比优化；
    第二收敛比优化子模块，设置为当所述实际带宽调整值不满足误差分析条件时，新增至少一个业务隧道组，计算增加后的所有业务隧道组的组合的实际带宽调整值，直到该实际带宽调整值满足误差分析条件。
  11. 一种端口优化系统，包括端口以及如权利要求7-10任一项所述的端口优化装置。
PCT/CN2017/078287 2016-05-27 2017-03-27 一种端口优化方法、装置及系统 WO2017202132A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610370091.7 2016-05-27
CN201610370091.7A CN107438013B (zh) 2016-05-27 2016-05-27 一种端口优化方法、装置及系统

Publications (1)

Publication Number Publication Date
WO2017202132A1 true WO2017202132A1 (zh) 2017-11-30

Family

ID=60412051

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/078287 WO2017202132A1 (zh) 2016-05-27 2017-03-27 一种端口优化方法、装置及系统

Country Status (2)

Country Link
CN (1) CN107438013B (zh)
WO (1) WO2017202132A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114826950B (zh) * 2022-05-30 2023-06-30 中国联合网络通信集团有限公司 端口数量预测方法、装置及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1770711A (zh) * 2005-11-18 2006-05-10 史文勇 一种瓶颈带宽和剩余带宽的测量方法和系统
CN101707788A (zh) * 2009-10-27 2010-05-12 北京邮电大学 基于差异化定价策略的多层网络业务动态规划方法
CN102130824A (zh) * 2010-10-30 2011-07-20 华为技术有限公司 联合优化的方法、装置和系统
CN103023812A (zh) * 2012-11-23 2013-04-03 上海寰创通信科技股份有限公司 一种多用户带宽均衡配置方法
US20160149829A1 (en) * 2014-11-25 2016-05-26 Vmware, Inc. Reverse breadth-first search method for optimizing network traffic in a distributed system with a point of convergence

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100508469C (zh) * 2006-12-21 2009-07-01 华为技术有限公司 一种在带宽按需分配业务中调整带宽的方法
CN101478495B (zh) * 2009-01-20 2011-08-10 华为技术有限公司 流量限制方法和装置
CN103297481A (zh) * 2012-03-05 2013-09-11 中兴通讯股份有限公司 互联交换方法和装置
CN104426801B (zh) * 2013-09-05 2018-03-23 中国移动通信集团广东有限公司 一种ptn网络的规划方法及装置
CN105591769A (zh) * 2014-10-22 2016-05-18 中兴通讯股份有限公司 一种业务端口的预留带宽配置方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1770711A (zh) * 2005-11-18 2006-05-10 史文勇 一种瓶颈带宽和剩余带宽的测量方法和系统
CN101707788A (zh) * 2009-10-27 2010-05-12 北京邮电大学 基于差异化定价策略的多层网络业务动态规划方法
CN102130824A (zh) * 2010-10-30 2011-07-20 华为技术有限公司 联合优化的方法、装置和系统
CN103023812A (zh) * 2012-11-23 2013-04-03 上海寰创通信科技股份有限公司 一种多用户带宽均衡配置方法
US20160149829A1 (en) * 2014-11-25 2016-05-26 Vmware, Inc. Reverse breadth-first search method for optimizing network traffic in a distributed system with a point of convergence

Also Published As

Publication number Publication date
CN107438013B (zh) 2022-05-17
CN107438013A (zh) 2017-12-05

Similar Documents

Publication Publication Date Title
WO2016127547A1 (zh) 一种虚拟机部署位置的调整方法、装置及系统
CN107729147A (zh) 流计算系统中的数据处理方法、控制节点及流计算系统
Guo et al. A cooperative game based allocation for sharing data center networks
Schaffrath et al. Optimizing long-lived cloudnets with migrations
CN104052681B (zh) 流量控制方法和装置
US7636316B2 (en) Resource allocation plan in a network
US20180279261A1 (en) Resource allocation device and resource allocation method
Draxler et al. Joint optimization of scaling and placement of virtual network services
WO2020004380A1 (ja) 割当装置、システム、タスク割当方法及びプログラム
CN108476175B (zh) 使用对偶变量的传送sdn流量工程方法与系统
CN105704054A (zh) 数据中心网络流量迁移方法及其系统
CN109325066A (zh) 一种数据库切换方法及终端设备
US11044161B2 (en) Method and system for latency-aware embedding of a virtual network onto a substrate optical network
Miao et al. Multicast virtual network mapping for supporting multiple description coding-based video applications
CN107872405A (zh) 分布式带宽分配和调节
CN112506650A (zh) 资源分配方法、系统、计算机设备和存储介质
CN105577572A (zh) 基于预算限制自组织云任务执行时间最短化资源分配方法
WO2017202132A1 (zh) 一种端口优化方法、装置及系统
US20150215230A1 (en) Methods and apparatus for allocating cloud-based media resources
CN103856385A (zh) 一种基于链路优先的虚拟网络映射方法
CN105760310B (zh) 地址分配方法及ddr控制器
CN109660623A (zh) 一种云服务资源的分配方法、装置和计算机可读存储介质
CN103338159B (zh) 轮询调度实现方法和装置
Wang et al. A virtual network embedding algorithm based on mapping tree
Chen et al. Resource distribution equilibrium for virtual network embedding over flexi-grid optical networks

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17801979

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17801979

Country of ref document: EP

Kind code of ref document: A1